Full text of "USPTO Patents Application 10687850"

See other formats

WORLD INTELLECTUAL PROPERTY ORGANIZATION
International Bureau

PCT

INTERNATIONAL APPLICATION PUBLISHED UNDER THE PATENT COOPERATION TREATY (PCT)

(51) International Patent Classification $ :
A61B 5/00, G01N 21/47

(11) International Publication Number: WO 99/59464

(43) International Publication Date: 25 November 1999 (25.1 1 .99)

(21) International Application Number: PCT/US99/10812

(22) International filing Date: 17 May 1999 (17.05.99)

(30) Priority Data:

09/080,470

18 May 1998 (18.05.98)

(71) Applicant: ABBOTT LABORATORIES [US/US];

CHAD-0377/AP6D-2, 100 Abbott Park Road t Ab-
bott Park, IL 60064-6050 (US).

(72) Inventors: KHALIL, Omar, S.; 1506 Portwine Court, Liber-

tyville, IL 60048 (US). DEMUL, Frits, F„ M.; De Delle 98,
NU7609 CJ Almelo (NL). HANNA, Charles, F.; 410 West
Lincoln Avenue, Libertyville, IL 60048 (US). STALDER,
Arnold, F4 3245 99th Street, Kenosha, WI 53142 (US).
YEH, Shu-Jen; 920 Stratford Court, Grayslake, IL 60030
(US). WU, Xiaomao; 17188 West Gurnee Glen, Gumee, IL
60031 (US). LOWREY, Michael, G.; Apartment 201, 33720
Royal Oak Lane, Wildwood, IL 60030 (US). KANGER, Jo-
hannes, S.; Westerstraat 62, NL-7546 BK Enschede (NL).
BOLT, Ren6, A.; Westerstraat 62, NL-7522 DB Enschede
(NL).

(74) Agents: WEINSTEIN, David, L. et al.; Abbott Laboratories,
CHAD 0377/AP6D-2, 100 Abbott Park Road, Abbott Park,
IL 60064-6050 (US).

(81) Designated States: GA, JP. European patent (AT, BE, CH, CY,
DE, DK, ES, FI, FR, GB, GR, IE, IT. LU, MC, NL, PT,
SE).

Published

With international search report.

(54) Title: NON-INVASIVE OPTICAL SENSOR WITH CONTROL OF TISSUE TEMPERATURE

DETECTOR 1

DETECTOR 3

SOURCE

DETECTOR 2

(57) Abstract

Devices and methods for
non-invasively measuring at least
one parameter of a sample, such
as the presence or concentration of
an analyte, in a body part wherein
the temperature is controlled.
The present invention measures
light that is reflected, scattered,
absorbed, or emitted by the sample
from an average sampling depth,
day, that is confined within a
temperature controlled region in
the tissue. This average sampling
depth is preferably less man 2
mm, and more preferably less than
1 mm. Confining the sampling
depth into the tissue is achieved
by appropriate selection of the
separation between the source and
the detector and the illumination
wavelengths. In another aspect,
the invention involves a method
and apparatus fox non-invasively
measuring at least one parameter
of a body part with temperature
stepping. In another aspect, the
invention involves a method and
apparatus for non-invasively
measuring at least one parameter

of a body part with temperature modulation. In another aspect, the invention provides an improved method of measuring at least one
parameter of a tissue sample comprising the steps of: (a) lowering the temperature of said tissue sample to a temperature that is lower
than the normal physiological temperature of die body; and (b) determining at least one optical property of said tissue sample.

dav3

FOR THE PURPOSES OP INFORMATION ONLY

Codes used to identify States party to the PCT on the front pages of pamphlets publishing international applications under the PCT.

Albania

Spain

Lesotho

Slovenia

Armenia

Finland

Lithuania

Slovakia

Austria

Prance

Luxembourg

Senega]

Australia

Gabon

Latvia

Swaziland

Azerbaijan

United Kingdom

Monaco

Chad

BA
BB

Bosnia and Herzegovina

Georgia

Republic of Moldova

Togo

Barbados

Ghana

Madagascar

Tajikistan
Turkmenistan

Belgium

Guinea

The former Yugoslav

Burkina Faso

Greece

Republic of Macedonia

Turkey

Bulgaria

Hungary

Mali

Trinidad and Tobago

Benin

Ireland

Mongolia

Ukraine

Brazil

Israel

Mauritania

Uganda .

Belarus

Iceland

Malawi

United States of America

Canada

Italy

Mexico

Uzbekistan

Central African Republic

Japan

Niger

Viet Nam

Congo

Kenya

Netherlands

Yugoslavia

Switzerland

Kyrgyzstan

Norway

Zimbabwe

C6te d'Tvoire

Democratic People's

New Zealand

Cameroon

Republic of Korea

Poland

Republic of Korea

Portugal
Romania

Cuba

Kazakstan

Czech Republic

Saint Lucia

Russian Federation

Germany

Liechtenstein

Sudan

Denmark

Sri Lanka

Sweden
Singapore

Estonia

Liberia

WO 99/59464

PCT/US99/10812

NON-INVASIVE OPTICAL SENSOR WITH CONTROL OF TISSUE TEMPERATURE

BACKGROUND OF THE INVENTION

1 . Field of the Invention

This invention relates to devices and methods for measuring the
concentration of one or more analytes in a human body part. More specifically,
this invention relates to devices and methods for the noninvasive determination
10 of in vivo analyte concentrations under conditions of precise temperature
control.

2. Discussion of the Art

15 Non-invasive optical monitoring of metabolites is an important tool for

clinical diagnostics. The ability to determine an analyte, or a disease state, in a
human subject without performing an invasive procedure, such as removing a
sample of blood or a biopsy specimen, has several advantages. These
advantages include ease of performing the test, reduced pain and discomfort,

20 and decreased exposure to potential biohazards. The result will be increased
frequency of testing, accurate monitoring and control, and improved patient care.
Representative examples of non-invasive measurements include pulse oximetry
for oxygen saturation (U. S. Patent Nos/3,638,640; 4,223,680; 5,007,423;
5,277,181; 5,297,548), laser Doppler flowmetry for diagnosis of circulation

25 disorder (Toke et ai, "Skin microvascular blood flow control in long duration

diabetics with and without complication", Diabetes Research, Vol. 5, Pages 189-
192, 1987), determination of tissue oxygenation (WO 92/20273), determination
of hemoglobin (U. S Patent No. 5,720,284) and of hematocrit (U. S Patent Nos.
5,553,615; 5,372,136; 5,499,627; WO 93/13706).

WO 99/59464 PCT/US99/10812

Measurements in the near-infrared spectral region are commonly
proposed, or used, in prior art technologies. The 600 - 1 1 00 nm region of the
spectrum represents a window between the visible hemoglobin and melanin
absorption bands and the infrared strong water absorption band. Light can
5 penetrate deep enough in the skin to allow use in a spectral measurement or a
therapeutic procedure.

Oximetry measurement is very important for critical patient care,
especially after use of anesthesia. Oxygenation measurements of tissue are
also important diagnostic tools for measuring oxygen content of the of the brain

10 of the newborn during and after delivery and for sports medicine and tissue
healing monitoring. Non-invasive determination of hemoglobin and hematocrit
would offer a simple non-biohazardous painless procedure for use in blood
donation centers, thereby increasing the number of donations by offering an
alternative to the invasive procedure, which is inaccurate and could lead to

15 rejection of a number of qualified donors. Hemoglobin and hematocrit values are
useful for the diagnosis of anemia in infants and mothers, without the pain
associated with pediatric blood sampling. Non-invasive determination of
hemoglobin has been studied in the art as a method for localizing tumors and
diagnosis of hematoma and internal bleeding. Non-invasive hematocrit

20 measurements can yield important diagnostic information on patients with kidney
failure before and during dialysis. There are more than 50 million dialysis
procedures performed in the United Stated and close to 80 million procedures
performed world-wide per year.

The most important potential advantage for non-invasive diagnostics

25 possibly will for non-invasive diagnosis of diabetes. Diabetes mellitus is a
chronic disorder of carbohydrate, fat, and protein metabolism characterized by
an absolute or relative insulin deficiency, hyperglycemia, and glycosuria. At least
two major variants of the disease have been identified. 'Type I" accounts for
about 1 0% of diabetics and is characterized by a severe insulin deficiency

30 resulting from a loss of insulin-secreting beta cells in the pancreas. The

WO 99/59464 PCT/US99/10812

remainder of diabetic patients suffer from 'Type II", which is characterized by an
impaired insulin response in the peripheral tissues (Robbins, S. L et al.,
Pathologic Basis of Disease . 3rd Edition. W. B. Saunders Company.
Philadelphia, 1984, p. 972). If uncontrolled, diabetes can result in a variety of

5 adverse clinical manifestations, including retinopathy, atherosclerosis,
microangiopathy, nephropathy, and neuropathy. In its advanced stages,
diabetes can cause blindness, coma, and ultimately death.

The principal treatment for Type I diabetes is periodic insulin injection.
Appropriate insulin administration can prevent, and even reverse, some of the

10 adverse clinical outcomes for Type I diabetics. Frequent adjustments of the
blood glucose level can be achieved either by discrete injections or, in severe
cases, via an implanted insulin pump or artificial pancreas. The amount and
frequency of insulin administration is determined by frequent or, preferably,
continuous testing of the level of glucose in blood (L e., blood glucose level).

15 Tight control of blood glucose in the "normal range", 60-120 mg/dL, is

necessary for diabetics to avoid or reduce complications resulting from
hypoglycemia and hyperglycemia. To achieve this level of control, the American
Diabetes Association recommends that diabetics test their blood glucose five
times per day. Thus, there is a need for accurate and frequent or, preferably,

20 continuous glucose monitoring to combat the effects of diabetes.

Conventional blood glucose measurements in a hospital or physician's
office rely on the withdrawal of a 5-10 mL blood sample from the patient for
analysis. This method is slow and painful and cannot be used for continuous
glucose monitoring. An additional problem for hospitals and physician offices is

25 the disposal of testing elements that are contaminated by blood.

Implantable biosensors have also been proposed for glucose
measurement. (G. S. Wilson, Y. Zhang, G. Reach, D. Moatti-Sirat, V. Poitout, D.
R. Thevenot, F. Lemonriier, and J.-C. Klein, Clin. Chem. 38, 1613 (1992)).
Biosensors are electrochemical devices having enzymes immobilized at the

30 surface of an electrochemical transducer.

WO 99/59464

PCT/US99/10812

Portable, "minimally-invasive" testing systems are now commercially
available. These systems require the patient to stick themselves to obtain a drop
of blood which is then applied to a disposable test strip containing coated
reagents or an electrochemical test element.
5 Although the portable instruments that read the strips are relatively

inexpensive ($100-$200), the cumulative cost to diabetics for the disposable
strips is considerable. Compliance is another major problem for minimally
invasive techniques. Finger sticks are painful and can result in infections,
scarring, and nerve damage in the finger. Disposal of potentially biohazardous

10 test strips and lancets is yet another problem with these systems.

"Non-invasive" (alternatively referred to herein as "Nl") glucose sensing
techniques measure in-vivo glucose concentrations without collecting a blood
sample. As defined herein, a "non-invasive" technique is one that can be used
without removing a sample from, or without inserting any instrumentation into,

15 the tissues. The concept involves irradiating a vascular region of the body with
electromagnetic radiation and measuring the spectral information that results
from one of four primary processes: reflection, absorption, scattering, and
emission. The extent to which each of these processes occurs is dependent
upon a variety of factors, including the wavelength and polarization state of the

20 incident radiation and the glucose concentration in the body part. Glucose
concentrations are determined from the spectral information by comparing the
measured spectra to a calibration curve or by reference to a physical model of
the tissue under examination. Various categories of non-invasive glucose
measurement techniques will now be described.

25 Nl techniques that utilize the absorption of infrared radiation can be

divided into three distinct wavelength regimes: Near-infrared (NIR), Mid-infrared
(MIR) and Far-infrared (FIR). As defined herein, NIR involves the wavelength
range from about 600 nm to about 1200 nm, MIR involves the wavelength range
from about 1200 nm to about 3000 nm and FIR involves the wavelength range

WO 99/59464

PCT/US99/10812

from about 3000 nm to about 25000 nm. As defined herein, "infrared" (or IR) is
taken to mean a range of wavelengths from about 600 nm to about 25000 nm.

U. S. Patent Nos. 5,086,229; 5,324,979; and 5,237,178 describe non-
invasive methods for measuring blood glucose level involving NIR radiation. In
5 general, a blood-containing body part (e. g., a finger) is illuminated by one or
more light sources, and the light that is transmitted through the body part is
detected by one or more detectors. A glucose level is derived from a
comparison to reference spectra for glucose and background interferants.
The 600-1100 nm spectral region contains a portion of the hemoglobin and water

10 absorption bands, which are several orders of magnitude more intense than
glucose overtone absorption bands. Thus, errors in the measurement of
hemoglobin absorption, water absorption, tissue scattering, and blood scattering
will greatly affect the glucose signal measured in this spectral range.
Determination of hemoglobin and study of the factors affecting the hemoglobin-

15 related signal are important for the determination of glucose when spectral data
generated in the NIR region are employed. Thus, in addition to the diagnostic
value of hemoglobin and hematocrit determinations, these determinations are
important for estimating the variability in non-invasive glucose measurements.
The NIR spectral region has been used for determination of blood oxygen

20 saturation, hemoglobin, hematocrit, and tissue fat content. It is also used for
exciting and detecting compounds in photodynamic therapy.

The use of MIR radiation for Nl glucose measurement has been described
in U. S. Patent Nos. 5,362,966; 5,237,178; 5,533,509; and 4,655,225. The
principles of operation are similar to those described for NIR radiation, except

25 that the penetration depth of the MIR radiation is less than that of NIR radiation.
As a consequence, most measurements in this region have been performed
using a backscattering geometry. As defined herein, a "backscattering
geometry" describes a configuration wherein scattered radiation is collected on
the same side of the sample as the entry point of the incident radiation. A

30 "transmission geometry" describes a configuration wherein light is transmitted

WO 99/59464

PCT/US99/10812

through the sample and collected on the side of the sample opposite to the entry
point of the incident radiation. This spectral region is less useful for the
determination of hemoglobin and hematocrit. However the 1300-1390 nm
wavelength has been used as a reference and water absorption wavelength for
5 hematocrit determination.

FIR measurements have been described in U. S. Patent Nos. 5,31 3,941 ;
5,115,133; 5,481,113; 5,452,716; 5,515,847; 5,348,003; and DE 4242083.

The photoacoustic effect results from the absorption of a pulse of optical
energy by tissues of a test subject, which optical energy is rapidly converted into
10 thermal energy. The subsequent thermal expansion generates an acoustic
pressure wave, which is measured by an acoustic transducer. In addition to the
absorption of light, the measured photoacoustic signal depends upon the speed
of sound in the medium, the thermal expansion coefficient, and the specific heat
of the medium.

15 Glucose measurements employing the photoacoustic effect have been

described by Quan et al. (K. M. Quan, G. B. Christison, H. A. MacKenzie, P.
Hodgson, Phys. Med. Biol., 38 (1993), pp. 1911-1922) and U.S. Patent No.
5,348,002.

Methods for the determination of glucose concentrations using changes in
20 the polarization of light are described WO 92/1 01 31 , WO 93/07801 , WO

94/02837, WO 94/05984, and WO 94/13199 and U. S. Patent Nos. 4,882,492;
5,086,229; 5,209,231 ; 5,21 8,207; 5,321 ,265; 5,337,745; 5,361 ,758; and
5,383,452.

An electromagnetic wave incident on an isolated molecule with an
25 electron cloud will cause the electrons to oscillate about their equilibrium

positions, in synchrony with the applied wave. The resulting electronic oscillator
instantaneously emits radiation (scatters) in all directions in a plane
perpendicular to the oscillating electrons. Most of the scattered photons are
elastically scattered,

WO 99/59464

PCT/US99/10812

i. e.. they have the same frequency as the incident radiation. A small fraction of
the scattered light (less than one in a thousand incident photons) is inelastically
(Raman) scattered. Unless otherwise indicated herein, "scattering" refers to
elastic scattering.

5 Because of the multiple scattering effect of tissue, optical measurements,

whether in transmission or reflectance, will contain tissue scattering information,
as well as absorption information. Tissue scattering information includes cell
size and cell shape, depth of layers and refractive index of intracellular fluids and
extracellular fluids. Absorption information includes absorption by visible

10 components, such as hemoglobin, melanin, and bilirubin, and the overtone
absorption of water, glucose, lipids, and other metabolites.

Spatially resolved light scattering (SRLS) techniques are a subset of the
elastic scattering methods previously described. As shown in FIG 1 , light is
injected into the surface of a tissue sample, such as a body part, at an injection

15 point. The diffusely reflected light, R, is measured at two or more detection
points located on the sample surface (e. g., the skin) at different detector
distances, r, from the injection point- The dependence of the intensity of the
diffuse reflectance R as a function of the detector distance (r) is used to derive
scattering and absorption coefficients of the tissue sample. These coefficients,

20 in turn, are related to the concentration of analyte(s). SRLS techniques have
been described U. S. Patent Nos. 5,551 ,422; 5,676,143; 5,492,1 1 8; 5,057,695,
European Patent Application EP 0810429, and in the journal literature (B.
Chance, H. Liu, T. Kitai, Y. Zhang, Analytical Biochemistry, 227, 1995, pp. 351-
362. H. Liu, B. Beauvoit, M. Kimura, B. Chance, Journal of Biomedical Optics,

25 1(2), April, 1996, pp. 200-21 1. J. Qu, B. Wilson, Journal of Biomedical Optics,
2(3), July 1997, pp. 319-325; A. Kienle, L Lilge, M. Patterson, R. Hibst, R.
Steiner, B. Wilson, Applied Optics, 35(13), May 1996, pp. 2304-2314.

Frequency-domain reflectance measurements use optical systems similar
to those used for spatially resolved light scattering (R as a function of r), except

30 that the light source and the detector are modulated at a high frequency (U. S.

.1 _ . I

WO 99/59464 PCT/US99/10812

Patent Nos. 5,187,672; 5,122,974). The difference in phase angle and
modulation between injected and reflected beam is used to calculate the
reduced scattering coefficient and the absorption coefficient of the tissue or
turbid medium. US Patent No. 5,492,769 describes frequency domain method
5 and apparatus for the determination of a change in the concentration of an
analyte, and U. S. Patent No. 5,492,1 18 describes a method and apparatus for
determination of the scattering coefficient of tissues.

U. S. Patent No. 5,553,616 describes the use of Raman scattering with
NIR excitation and an artificial neural network for measuring blood glucose level.

10 Although glucose Raman bands are distinct from protein Raman bands,

sensitivity of this method limits its applicability for in-vivo measurements. WO
92/10131 discusses the application of stimulated Raman spectroscopy for
detecting the presence of glucose.

The Nl techniques described above are painless, reagentless, and are

15 expected to be less expensive than the finger stick approach over the long term
use by a patient. Nl techniques also eliminate the potentially biohazardous
waste associated with invasive and minimally invasive measurements. However,
Nl methods have not yet achieved the level of accuracy and precision that is
required for measuring physiologically relevant concentrations of glucose in-vivo.

20 A major challenge for all of the non-invasive techniques to date has been

to collect spectral information with sufficiently high signal-to-noise ratios to
discriminate weak glucose signals from the background noise. In the ideal case,
a non-invasive sensor would be highly sensitive for the parameter of interest
(e, g., glucose concentration) while remaining insensitive to interfering analytes

25 or physiological parameters. In practice, all of the non-invasive measurement
techniques described in the prior art are sensitive to one or more interfering
"physiological" or "spectral" variables.

As used herein, the expression "physiological variables" describes
physiological parameters, such as temperature, that can adversely affect the

30 sensitivity or selectivity of a non-invasive measurement. As used herein, the

WO 99/59464

PCT/US99/108I2

expression "spectral variables" describes spectral features that arise either from
poorly resolved analyte bands or from other interfering components in the
sample. Several significant sources of spectral interference for the Nl
determination of glucose in biological samples are water/ hemoglobin, albumin,

5 cholesterol, urea, etc. Other tissue constituents that are present at lower
concentrations or have lower absorption or scattering cross-sections may also
contribute to an overall background signal that is difficult to separate.

Physiological and spectral variables can introduce unwanted noise, or
worse, completely overwhelm the measured signals of interest (e. g., those

10 related to glucose concentration). It is difficult to eliminate these interferences
because they may exhibit one or more of the following properties:

(a) they may contribute nonlinearly to the measured signal,

(b) they may vary with spatial location within the sample,
15 (c) they may vary over time, or

(d) they may vary from sample to sample.

Co-pending U. S. Application Serial No. 08/982,939, filed December 2,
1997, assigned to the assignee of this application, describes a multiplex sensor
20 that combines at least two Nl techniques selected from those described above
in order to compensate for the effects of spectral and physiological variables. A
description of prior art measurements in which tissue temperature is controlled is
provided below.

U.S. PatentNos. 3,628,525; 4,259,963; 4,432,365; 4,890,619; 4,926,867;

25 5,131,391, and European Patent Application EP 0472216 describe oximetry
probes with heating elements that are placed against a body part. These
devices enhance sensitivity of the oximeter by elevating local tissue perfusion
rates, thereby increasing hemoglobin concentrations. U. S. Patent No.
5,148,082 describes a method for increasing the blood flow in a patient's tissue ,

30 during a photoplethsmography measurement, by warming the tissue with heat

WO 99/59464

PCT/US99/I0812

generated by a semiconductor device mounted in a sensor. The heating
element comprises a less efficient photodiode that acts as a heat source and as
a light source.

U. S. Patent No. 5,551 ,422 describes a glucose sensor that is "brought to
5 a specified temperature preferably somewhat above normal body temperature
(above 37° C) with a thermostatically controlled heating system". Unlike the
oximetry sensors, simply increasing tissue perfusion without controlling it is
contraindicated for glucose measurements, because hemoglobin interferes with
glucose measurement. This patent also fails to account for large variations in

10 scattering intensity that result from the temperature gradient between the skin
surface and the interior of the body part. As will be described more thoroughly
below, the smallest devices disclosed in that patent have an average sampling
depth of 1 .7 mm. Depths and lateral distances of several millimeters are
sampled at the longest spacings between source and detector taught in that

15 patent. As shown in FIG. 1 and as defined herein, the average sampling depth,
d a v» is the average penetration depth along an axis normal to the tissue surface
that is sampled in a given Nl measurement. A thermal model of the human
forearm, shown in FIGS. 6-8, suggests that, depending on the ambient
temperature, the temperature of the tissue at a depth of 1.7 mm could be as

20 much as 0.5° C warmer than that of the skin surface. According to Wilson et aL,
(J. Qu, B. Wilson, Journal of Biomedical Optics, 2(3), July 1997, pp. 319-325),
the change in scattering expected for a 0.5° C change in temperature is
equivalent to a 5 mM (90 mg/dL) change in glucose concentration. Thus, the
scattering variability due to the temperature gradient probed by U. S. Patent No.

25 5,551,422 is as large as the signal expected for normal physiological glucose
levels.

Although a variety of spectroscopic techniques are disclosed in the prior
art, there is still no commercially available device that provides noninvasive
glucose measurements with an accuracy that is comparable to invasive

WO 99/59464 PCTAJS99/10812

methods. All of the prior art methods respond to glucose concentrations, but
they are also sensitive to physiological and spectral variables. As a result,
current approaches to non-invasive glucose testing have not achieved
acceptable precision and accuracy.

5 Thus, there is a continuing need for improved Nl instruments and methods

that are unaffected by variations in tissue such as temperature and perfusion.
There is also a need for improved Nl instruments and methods that will provide
essentially the same accuracy as conventional, invasive blood glucose tests.
There is also a need for low-cost, reagent-free, painless, and environmentally

10 friendly instruments and methods for measuring blood glucose levels in diabetic
or hypoglycemic patients.

SUMMARY OF THE INVENTION

In one aspect, the present invention involves devices and methods for
non-invasively measuring at least one parameter of a sample, such as the
presence or concentration of an analyte, in a body part wherein the temperature
is controlled. As will be described more fully below, the present invention

20 measures light that is reflected, scattered, absorbed, or emitted by the sample
from an average sampling depth, d^, that is confined within a temperature
controlled region in the tissue. This average sampling depth is preferably less
than 2 mm, and more preferably less than 1 mm. Confining the sampling depth
into the tissue is achieved by appropriate selection of the separation between the

25 source and the detector and the illumination wavelengths.

Confining the sampling depth provides several advantages. First, the
entire signal is acquired from a region of tissue having a substantially uniform
temperature. As defined herein, a "substantially uniform tissue temperature"
means that the temperature of the tissue varies by no more than ±0.2° C,

30 preferably no more than +0.1° C. Secondly, the sampled tissue region is more

.11

WO 99/59464

PCT/US99/10812

homogeneous than the tissue regions sampled by the devices of the prior art.
As a result, physiological and spectral interferences are controlled so that their
contributions may be separated, quantified, and removed from the signals of
interest.

In the present invention, the area of the skin of the body part where
temperature is controlled is larger than the optical sampling area. A preferred
ratio of the area of controlled temperature (surface area of the temperature
controlled body interface) to the optical sampling area (surface area of the optical
probe) is greater than 2:1, preferably greater than 5:1 .

In another aspect, the present invention involves a method and apparatus
for non-invasively measuring at least one parameter of a body part with
temperature stepping. As defined herein, "temperature stepping" comprises
changing the temperature of a tissue sample between at least two different
predefined temperatures. Non-invasive measurements are performed at each of
the two or more different temperatures in order to remove the effects of
temperature fluctuations on the measurement.

In another aspect, the present invention involves a method and apparatus
for non-invasively measuring at least one parameter of a body part with
temperature modulation. As used herein, temperature modulation consists of
cycling the temperature (changing the temperature repeatedly) between at least
two different predefined temperatures. Non-invasive measurements are
performed at each of the two or more different temperatures in order to eliminate
the effects of temperature fluctuations on the measurement.

In another aspect, the present invention provides an improved method of
measuring at least one parameter of a tisisue sample comprising the steps of:

(a) lowering the temperature of said tissue sample to a temperature that is
lower than the normal physiological temperature of the body; and

(b) determining at least one optical property of said tissue sample.

WO 99/59464

PCT/US99/10812

In another aspect, the present invention provides a method of measuring
at least one parameter of a tissue sample comprising the steps of:

(a) stepping the temperature of said tissue sample between at least two
5 different temperatures;

(b) measuring said at least one optical property of the tissue sample as a
function of temperature;

10 (d) correlating the at least one parameter of the tissue sample with the

functional dependence of the at least one optical property on temperature.

The present invention is particularly advantageous for biological samples
where multiple interfering analytes or physiological variables can affect the

15 measurement. Non-invasive measurements may be made on a body part of a
patient, e. g., a finger, earlobe, lip, toe, skin fold, or bridge of the nose.

The invention offers several advantages over the prior art.
At small separations of source from detector, tight samples penetrate the tissue
to a lower depth, where smaller temperature gradients are encountered, than to

20 deeper regions of the tissue. In addition, better temperature control can be
achieved at lower depths of penetration in the sampled region. If the separation
of source from detector varies over large distances (e. g M 0.5 cm - 7 cm), light
from the source and detected light propagates through the epidermis, the
dermis, as weli as deeper regions of tissue, including the subcutis (which has

25 higher fatty adipose tissue content) arid underlyirig muscle structures. These
layers provide sources of variability in measurements because of the difference
in cell size, cell packing, blood content, as well as thermal properties.

In addition, for tissue that is heterogeneous along dimensions parallel to
the skin surface (x and y), there is lower likelihood of photons encountering

30 tissue components that will cause anomalies in the scattering measurements. It

WO 99/59464

PCT/US99/10812

is also possible to perform measurements on a small localized area of the skin
with a probe design having a closely spaced source and detector than with a
source that is located a great distance from the detector. Thus, it is possible to
detect blood vessels and hair fibers and determine their effect on the signal.

5 Probes having large separations of source from detector require the use

of a large body mass, such as the muscle of the arm, thigh, or the abdomen.
Accordingly, the body site locations where such a probe can be used on are
limited, and substantial disrobing and inconvenience for the user is required.
Thus, another advantage of the probe design of the present invention is that

10 probes of 5 mm or less can be used, particularly with small body parts, such as
ear lobes and fingers. However, probes of 5 mm or less can also be used with
larger body parts, such as the forearm, thigh, or abdomen.

Another advantage of a small separation between source and detector is
the higher signal to noise ratio obtainable at small separations due to increases

15 in the amount of light reaching the detector. Thus simpler, inexpensive, rugged
components, such as light emitting diodes, small flash lamps, and incandescent .
lamps, can be used as light sources, and commercially available inexpensive
photodiodes can be used as detectors. Probes having a large separation
between source and detector use laser diodes and photomultiplier tubes,

20 because weaker signals are generated.

In addition to convenience and cost advantages, other engineering design
considerations favor the probe design of the present invention. It is preferred to
generate a constant temperature using standard Peltier cooler elements that are
approximately 1 cm squares. In order to obtain an aspect ratio of 5/1 , probes of

25 2 mm or less are desirable, especially for use with small body parts, such as ear
lobes and fingers. Larger thermoelectric cooling and heating elements may be
employed at a cost of higher power consumption and greater heat dissipation.

Prior art measurements that use separations of detector and source in
excess of 3 mm result in the phase and polarization of the incident light that are

30 randomized. However, in the present invention, the preferred separations are

WO 99/59464

PCT/US99/10812

less than 2 mm, and polarization and interference effects can be measured. The
use of polarizers and polarization conserving fibers can reveal some internal
sample properties. In addition, temperature effects on transmission of polarized
light through tissue can be studied with the appiaratus of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic diagram of the average sampling depth, d av , for a
10 given spatially-resolved light scattering measurement.

FIG. 2 is a schematic diagram of the temperature-controlled
backscattering system of the present invention.

FIG. 3A is a diagram illustrating the bifurcated optical fiber probe of FIG.
2. FIG. 3B is a series of diagrams showing portions of the bifurcated optical
15 probe of FIG. 3A.

FIG. 4 is a diagram illustrating the nominal separation distances, r,
between collection, fibers 2-7 and the excitation fiber 1 .

FIG. 5 is a schematic diagram of the human interface module of the
temperature-controllable backscattering system of the present invention.
20 FIG. 6 is a schematic diagram of a thermal model of a human forearm.

FIGS. 7a, 7b, and 7c are schematic diagrams illustrating the temperature
gradients as a function of penetration depth and lateral distance results of the
thermal model of a human forearm,

FIG. 8 is a graph illustrating the temperature gradients over time predicted
25 by a thermal model of the tissues adjacent to temperature-controllable
backscattering system of the present invention.

FIG. 9 is a graph illustrating a Monte Carlo simulation and the measured
reflectance for human volunteers.

FIG. 10 is a graph illustrating spectral distribution of spatially resolved
30 scattering data from a diabetic subject and a non-diabetic subject

WO 99/59464 PCT/US99/10812

FIG. 1 1 is a graph illustrating spatially resolved scattering data, spectral
distribution of the partial change in the scattering coefficient with respect to
temperature, from a diabetic subject and a non-diabetic subject.

FIG. 12 is a graph illustrating spatially resolved scattering data, spectral
5 distribution of the change in absorption coefficient with respect to temperature,
from a diabetic subject and a non-diabetic subject.

DETAILED DESCRIPTION

As used herein, the expression "tissue optics" refers to the study of light
propagation in biological tissues. The expression "optical properties" refers to
the absorption, scattering, and depolarization properties of the tissues. The
expression "scattering media" refers to media that both scatter light and absorb

15 light. The expression "absorption coefficient, n a " refers to the probability of light
absorption per unit path length. The expression "scattering coefficient, m" refers
to the probability of light scattering per unit path length. The expression "isotropy
factor, g" refers to the average cosine of the scattering angle for a multiply
scattering photon. The expression "reduced scattering coefficient u s '" refers to

20 the probability of equivalents isotropic scattering per unit path length. The
reduced scattering coefficient is related to the scattering coefficient m and the
anisotropy factor g by the relationship u,' = (1-g) The expression "transport
optical mean free path" refers to the mean path length between photon-medium
interaction, which can be either absorption or scattering; mean free path = (1/(u,

25 + Ms'))- The expression "effective scattering coefficient" refers to the transport
attenuation coefficient, n^= ^^d^+y^y The expression "penetration depth. 6"
refers to the speed of light intensity decay in turbid media. Penetration depth is
determined by both the absorption and scattering coefficient, 8 =1/^ . The
expression "Monte Carlo simulations" refers to a statistical method that can be

WO 99/59464

PCT/US99/10812

used to trace photon propagation in turbid media. The expression "diffuse
reflectance" refers to a process by which light reflected from a boundary of a
sample is measured at all angles and over an area wider than the beam
diameter. The expressions "spatially resolved scattering" or "spatially resolved

5 reflectance" refers to a process by which light injected at certain point on a

boundary of the sample is detected at several light measurement points and at a
predetermined spacing from the light injection point. Alternatively, it can be
defined as the light detected at a given point on the sample boundary as a result
of injecting light at discrete points located on the same boundary at

10 predetermined separation distances. The expression "frequency domain

measurement" refers to a measurement of light involving the phase angle and/or
the amplitude change of a modulated light beam, at a given separation of source
and detector, as the beam transverses a scattering medium.

In "warm-blooded" animals, such as birds and mammals, a group of reflex

15 responses operate to maintain body temperature within a narrow range in spite
of wide fluctuations in environmental temperature. In humans, the normal value
for the oral temperature is 37° C, however, this temperature varies by about ±1°
C from individual to individual due to differences in metabolic rate, age, and
hormonal influences. The normal human core temperature undergoes a regular

20 circadian fluctuation of 0.5-0.7° C. In individuals who sleep at night and are
awake during the day, the temperature is lowest during sleep, slightly higher in
the awake but relaxed state, and rises with activity. In women, there is an
additional monthly cycle of temperature variation characterized by a rise in basal
temperature at the time of ovulation. Temperature regulation is less precise in

25 young children, and they may normally have a core temperature that is 0.5° C or
so above the established norm for adults.

Various parts of the body are at different temperatures, and the
magnitude of the temperature difference between the parts varies with the
environmental temperature. The rectal temperature is representative of the

30 temperature at the core of the body and varies least with changes in

WO 99/59464

PCT/US99/10812

environmental temperature. The extremities are generally cooler than the rest of
the body and, within a particular body part, the tissue temperature is lowest at
the skin surface.

Variations in tissue temperature affect other physiological variables, such
5 as the perfusion rate. A rise in tissue temperature triggers a homeostatic reflex,

which enhances local blood flow in order to increase transfer of heat away from

the skin. Cooling the tissue to approximately 25° C decreases the perfusion rate;

however, at much lower temperatures the skin again takes on a ruddy color.

Other factors, such as activity, infections, some malignancies or mental stress,
10 can also modulate the perfusion rate. A familiar example is the change in skin

coloration, which can accompany exercise, alcohol intake, or even a change in

position from sitting to standing.

In order to fully appreciate the effects of temperature variations on Nl

measurements, it is helpful to review the theoretical description of light
15 propagation in tissues. A discussion of optical properties of tissue and the effect

of these properties on light scattering and absorption is provided below. The

dependence of Nl measurements on temperature of the tissue is also illustrated,

and preferred embodiments for controlling the temperature of Nl measurements

are described.

20 For clear or highly absorbing samples, Beer's law describes the light

fluence within a sample as follows:

l = l 0 exp-(u,z) (i)

25 where I represents the light fluence at a distance, z, into the sample, l 0

represents the incident intensity and u, represents a total attenuation coefficient
Ht is the sum of the absorption coefficient, m, and the scattering coefficient, m.
The mean free path of a photon describes the average distance traveled by a
photon between absorptive or scattering events and is defined as 1/m .

WO 99/59464

PCT/US99/10812

At visible and NIR wavelengths, scattering dominates absorption in
biological tissues (i.e., m» ji a ), and photon propagation deviates significantly
from Beer's law. Tissue scattering occurs because of a mismatch between the
index of refraction of the extracellular fluid (ECF) or intracellular fluid (ICF) and

5 the cellular membranes of the tissue. As used herein, the expression "cellular
membranes" encompasses both the cell membrane as well as the membranes of
organelles, such as mitochondria or collagen fibrils. Besides undergoing
scattering and absorption, photons can be reflected at the tissue/air interface;
photons can also be re-emitted out of the tissue.

10 An exact assessment of light propagation in tissues would require a model

that characterizes the spatial and size distributions of tissue structures, their
absorbing properties, and their refractive indices. For real tissues, such as skin,
the task of creating a precise representation of photon migration from the
solution of Maxwell's electromagnetic (EM) wave equations is formidable.

15 Consequently, it is necessary to rely upon mathematical approximations in order
to simplify the calculation of optical properties of tissue.

One useful approach for describing the transfer of light energy through a
turbid medium uses radiative transport (RT) theory. In the RT formalism, light
propagation is considered equivalent to the flow of discrete photons, which may

20 be locally absorbed by the medium or scattered by the medium. For dense

media where the detector distance is large relative to the photon mean free path,
RT theory can be simplified to yield the Diffusion Theory (DT) approximation. DT
describes photon propagation in tissues by the absorption coefficient, m, and the
reduced scattering coefficient m' = m[l-g], where the anisotropy factor, g,

25 represents the average cosine of the angle at which a photon is scattered.
Typical values of g for tissues are 0.9<g<1.0 (forward scattering). The
attenuation of photons in tissues is described by an effective attenuation
coefficient, as follows:

WO 99/59464

PCT/US99/108I2

Har = V(3 p a (p a + p,') = V(3 p a [ Ma + p s (l-g)]) (2)

The value of M.* can be calculated from scattering measurements (such as
SRLS) and both p a and p $ ' can be derived from measurements of p^ under
different conditions. In turn, changes in the values of u a and p $ ' can be related to
tissue parameters, such as the concentration of an analyte.

For tissue samples irradiated at visible and NIR wavelengths, the size of
the scattering material is near the wavelength of light, and the reduced scattering
coefficient, p s ', can be expressed using Mie theory as follows:

Ms' = M»(l-g) = 3.28jia 2 p(27ran 4 A) 0J7 (m-l) JW (3)

where p represents the volume density, number of particles per unit volume, a
represents the radius of the scattering particle (e. g., cells, mitochondria, or
collagen fibrils), n ex represents the refractive index of the medium (ECF or ICF),
and m = (n^), which is the ratio of the refractive index of the scattering
particle n to to n„ . See Graaf, et al., " Reduced light-scattering properties for
mixtures of spherical particles: a simple approximation derived from Mie
calculations", APPLIED OPTICS, Vol. 31, No. 10, 1 April 1992. Light fluence
within the sample is described by the following formula:

I s | 0 exp-(MeflZ) (4)

where I, l 0 ,and z are defined as above and Me* is defined as above and differs
from the total p, defined in Equation (1).

For a given incident wavelength, p,' changes with either the cell size, a, or
the refractive index ratio m, as shown in equation 3; Because the refractive
index of the cellular membranes, remains relatively constant, u,' is influenced

WO 99/59464 PCT/US99/108I2

mostly by and a. For example, glucose reduces tissue scattering by
decreasing the refractive index difference between the ICF/ECF and the cellular
membranes. Variations in n^are not specific for a particular analyte, however,
and are affected by any change in the total concentration of solutes in the ECF,
5 including hemoglobin. n w is also susceptible to changes in physiological
variables, such as temperature of the tissue.

Methods of determining ^ m', and ji a are known in the art. One of these
methods is the measurement of diffuse reflectance of the skin tissue. In a
diffuse reflectance measurement, the measured reflectance has the following
10 functional dependence:

Rd = /0C**a. rVno)

where m' represents the reduced scattering coefficient, m represents the
15 absorption coefficient, n s represents the refractive index of the scattering medium

and n 0 represents the refractive index of the surrounding layer, usually air.

Another method of measuring the absorption and scattering coefficients is

known as spatially resolved diffuse reflectance, R(r). In this method, the intensity

of the reflected light is measured at several distances from the point at which
20 light is injected. The intensity of the reflected light at a given distance R(r ) is

related to the separation of the source and detector by the relationship:

R(r) = Ko[exp(^yr

25 A plot of Log r times R(r) vs. r yields a line with a slope of .

Other methods for determination of determination optical properties of tissues
are described in the art. These methods include collimated transmittance and
frequency domain measurements.

WO 99/59464

PCT/US99/10812

The present invention involves methods and devices for eliminating the
effects of physiological and spectral variables on measurements of n a and \x,\
Specific descriptions of the mechanisms by which temperature affects these
parameters are provided below.
5 Temperature fluctuations affect the sensitivity and selectivity of Nl

measurements by influencing physiological and spectral variables. For example,
temperature affects perfusion rates, which alter the concentrations of blood-
borne spectral variables, such as hemoglobin, water, and electrolytes These
spectral variables can introduce measurement inaccuracies. However, it has

1 0 been found that the magnitude of these inaccuracies can be reduced by
compensating factors. IR backscattering measurements may be used to
measure hemoglobin, total protein, HDL, and triglycerides in tissues.
Temperature-controlled IR measurements thus yield a more accurate measure of
glucose concentration by increasing the measurement accuracy of important

15 spectral variables.

Temperature also affects the refractive index of the ECF. Because
approximately 90% of human tissue is water, the refractive index of ECF may be
approximated by the refractive index of water which varies with temperature as:

20 n = 1.3341 +2.5185 10 s T- 3.6127 1oV + 2.3707 10* T*

where n represents the refractive index and T represents the temperature.
Changes in osmolarity of the ECF can also change the size of the cells due to
osmotic swelling and shrinkage.

Because water is the main constituent of biomedical samples, its optical
properties (in particular its absorption coefficient) are important parameters for Nl
measurements. Temperature variations contribute to the background noise in Nl
measurements by altering the intensities as well as the frequencies of the
30 dominant water absorption bands. Changes in the absorption and scattering

WO 99/59464

PCT/US99/10812

properties of the sample produce a variable optical path length, which affects all
Nl measurements.

Depolarization is a process in which completely polarized light is coupled
5 to unpolarized light and is defined as

D = Polarized Light
Total Incident Light

10 In turbid media, an incident polarized light beam undergoes multiple scattering
events. The polarization of the incident beam is degraded with each scattering
event, and the depolarization is affected by the number of scattering events in
the medium. Because temperature influences the overall refractive index, the
number of scattering interactions changes with varying tissue temperature. As

15 the number of scattering interactions increases, the polarized light becomes
progressively depolarized.

The following non-limiting examples further illustrate the present invention.

20 EXAMPLES

EXAMPLE 1

FIG. 2 is a schematic diagram of one embodiment of the temperature-
25 controlled backscatter system 10 (TCBS) of the present invention. The TCBS
comprises three modules: a human interface module 12; a light source module
14; and a detector module 16. As shown in FIG. 2, the human interface module
12 is connected to the light source module 14 and the detector module 16 via a
bifurcated optical fiber probe 18.
30 FIG. 3A is a detailed illustration of the bifurcated optical fiber probe 18.

The bifurcated optical fiber probe is constructed from Anhydrous G Low OH VIS-

WO 99/59464

PCT/US99/10812

NIR optical fibers. The bifurcated optical probe 18 comprises a source tip 19, a
detector tip 20, and a common tip 21 . The three distinct termination points or
"tips" of the bifurcated optical probe are shown in FIG. 3B. During operation, the
source tip 19 Is contained within the light source module 14, the detector tip 20 is
contained within the detector module 16, and the common tip 21 is contained
within the human interface module 12. A single optical fiber 22 transmits light
from the source tip 19 to the common tip 21. Six optical fibers 23, 24, 25, 26, 27,
and 28 transmit light from the common tip 21 to the detector tip 20.

Light source module 14 includes a source of modulated light (not shown),
such as a Gilway L1041 lamp modulated with a Stanford Research Optical
Chopper. A prism, a dichroic beam splitter, or the like may be used to direct a
portion of the beam emanating from the light source to a reference detector,
such as a Hammamatsu S-2386-44K 6C Silicon Detector, in order to normalize
the measurements for fluctuations in source intensity. The rest of the light
emanating from the light source is focused onto the end of the source tip by
means of at least one focusing lens. Additional optical elements, such as
attenuators, optical filters, and irises may be inserted between the light source
and the source tip. The source tip is preferably held in an adapter having
provisions for adjusting the location of the source tip with respect to the beam
emanating from the light source.

The common tip 21 is installed in the human interface module, which is
placed against a body part during use: As shown in FIG. 3B, the common tip
comprises the source fiber 22 and six additional fibers 23, 24, 25, 26, 27, and 28
that collect the light that is scattered by the tissue sample.

The collection fibers 23, 24, 25. 26, 27, and 28 are located within the
common tip 21 at increasing distances from the source fiber 22. The nominal
separation distances, r, between the center of the source fiber 22 and the
centers of the collection fibers 23, 24, 25, 26, 27, and 28 of the common tip 21
are shown in FIG. 4. An important aspect of the present invention is that all of
the collection fibers are located at separation distances, r, that are less than 4

WO 99/59464

PCT/US99/108I2

mm away, and, preferably, less than 2 mm away from the source fiber 22. As
will be more thoroughly described below, locating the fibers in this manner
results in enhanced precision and accuracy over the methods used in the prior
art.

5 The collection fibers 23, 24, 25, 26, 27, and 28 are arranged in a circle

within the detector tip 20, as shown in FIG. 3B, with sufficient spacing to allow a
shutter to interrogate each fiber individually. The detector module receives the
detector tip 20 and holds it adjacent to a rotating shutter (not shown) that allows
detection of the light emitted from one fiber at a time. The shutter has a detent

10 or other means to lock it in the six fiber positions. The light from the fiber of
interest is focused on a detector by a pair of 25 mm diameter, 60 mm focal
length Achromatic lenses. The detector was a Hammamatsu S-2386-44K 6C
Silicon Detector or any other equivalent detector. The detector module also
comprises appropriate electronic signal processing instrumentation such as large

15 dynamic range amplifiers and lock-in amplifiers. Alternatively, the outputs of the
six fibers can be directed to six detectors for parallel signal processing.

FIG. 5 illustrates the human interface module 12, which comprises an
aluminum disk 30, a thermoelectric cooling element 31, a thermocouple 32, a
heat sink 34, the common tip 21 , and an interface adapter 36. The aluminum

20 disk contains a through-hole that receives the common tip 21 of the fiber optic
probe and holds the common tip 21 against the body part. The temperature of
the aluminum disk 30 (and of the tissue adjacent the disk 30) is controlled by a
thermoelectric cooling element 31 , such as a Marlow Industries model number
SP1 507-01 AC. The thermoelectric cooling element 31 is powered by a

25 temperature controller/power supply, such as a Marlow Industries model number
SE5000-02.. A heat sink 34 is provided on the back of the thermoelectric cooling
element 31 to enhance heat transfer The interface adapter 36 is shaped to
conform to a body part and may, for example, be cylindrical, flat, spheroidal or
any other shape that provides efficient optical and thermal coupling to a body

30 part

WO 99/59464

PCT/US99/10812

EXAMPLE 2

FIG. 6 illustrates assumptions made for a mathematical model of the
human interface module and the underlying tissue sample. The tissue sample is
modeled as a multi-layer sample wherein each of the layers has different thermal
transfer properties. In the absence of the interface module, the tissue
temperature changes along a gradient from an interior body physiological
temperature of approximately 37° C (muscle) to a typical skin surface
temperature of 32.4° C (epidermis).

FIGS. 7a through 7c show model values of steady-state temperatures
within an axisymmetric cross-section of the model. Each figure represents a
different temperature condition imposed by the interface module (i. e., the disk).
The figures also place a dimensional scale along the edges of the model to
indicate values of tissue depth (up to 10 mm from the tissue surface) and radial
distance (up to 30 mm from the central axis of the disk).

FIG. 7a presents the situation when a disk made of an insulating material
such as Plexiglas is brought in contact to the tissue surface. The disk is not
capable of controlling temperature. The disk insulates the underlying tissue from
its ambient surroundings and causes a minor increase in epidermal temperature
from 32.4° C to about 32.7° C. At a depth of 2 mm below the tissue surface, the
temperature is about 33.0° C on the center axis. The resulting tissue
temperature gradient 0.6° C is somewhat smaller than the natural state wherein
no insulating disk is applied. As shown by Quan and Wilson, a 0.5° C
temperature difference affects the scattering signal by an amount equivalent to
90 mg/dL change in glucose concentration. This represents a model for an
insulating detection probe having no temperature control. The insulating probe
head model offers a minor advantage for the non-invasive measurement of an
analyte.

WO 99/59464

PCT/US99/10812

In FIG. 7b a heat conducting disk 30 maintained at a constant
temperature of 34.4° C is applied to the surface of the tissue. The disk maintains
the temperature of the epidermis at 34.4° C, which temperature is lower than the
body physiological temperature. At a depth of 2 mm below the surface of the

5 tissue, the temperature is 34.3° C. The disk produces a tissue temperature
gradient of 0.1° C, which is substantially smaller than the natural state. At a
depth of 2 mm, temperature is maintained within a 0.1° C at lateral distances up
to 5 mm from the axis. At a depth of 1 mm, temperature is maintained within
0.1 ° C at lateral distances up to 8 mm from the axis. The time required for the

10 tissue to reach this condition (after the skin is brought in contact with the disk) is
shown in FIG. 8. In FIG. 8, the model temperatures along the central axis just
prior to the moment the skin touches the heat conducting disk is shown and is
labeled as Time = 0. In the time period ranging from 30 seconds to 600 seconds
after contact between skin and the heat conducting disk, the tissue (dermis)

1 5 temperature rises and the temperature gradient decreases until an essentially
constant value is achieved. The calculated temperature gradient reaches 0.2° C
at a depth of 2 mm, after a 60 second equilibration period. At 600 seconds, the
temperature gradient flattens out and approaches the steady-state value labeled
as Time = infinity.

20 In FIG. 7c a heat conducting disk having a constant temperature of 30.4°

C is brought in contact with the tissue surface. Interaction with the disk reduces
the epidermal temperature to 30.4° C. Temperature is calculated to be 30.8° C
at a depth of approximately 1 mm and 31 .5° C at a depth of 2 mm below the
surface of the tissue. The temperature gradient is 0.4 C at a depth of 1 mm and

25 1 .1° C at a depth of 2 mm. This gradient is greater than the gradients shown in
FIGS. 7a and 7b. However, temperature was calculated to be constant in the
horizontal plane up to distance of 5 mm.

The probe design described herein is particularly well adapted for
temperature-controlled SRLS measurements. When the probe temperature is

30 held at 34.4° C (i. e., 2° C above the natural temperature of the surface of the

WO 99/59464

PCT/US99/10812

skin), the underlying tissue temperature does not vary by more than 0.2° C up to
a depth of 2 mm below the surface of the tissue (except at the extreme edges of
the disk, where no SRLS measurements are taken). Furthermore, the tissue
temperature does not vary by more than 0.1° C within a depth of 2 mm and a
radial distance of less than 4 mm from the central axis of the disk. Closer control
of temperature and less of a temperature gradient is maintained at 0.1 mm below
the surface of the skin. Greater penetration depths would encounter a volume of
tissue having a greater range of temperatures, thereby decreasing the
reproducibility of the SRLS measurement.

Photon path in a turbid medium can be expressed by the radiation
transport equation. This analytical equation is difficult to solve. An
approximation for solving the equation is the diffusion theory approximation. The
diffusion theory approximation is limited to cases where the light has been highly
scattered (i. e., the approximation is limited to situations in which a photon is
scattered many times before it is absorbed or detected). The condition of
multiple scattering depends upon the average distance between scattering
centers (density of scattering material) and on the ratio of scattering to total
attenuation known as optical albedo GVm + n 8 ). The source-tissue-detector
geometry and the boundary conditions of the medium are important for the
application of the diffusion theory approximation. For SRLS measurements, the
conditions require that the separation between the source and the detector be
much greater than the transport optical mean free path, the mean distance
between two successive absorption or scattering interactions in the medium
I1'(n 8 + Ms')]- For tissues with small absorption coefficients (1 cm 1 ) and a
scattering coefficient of 10 cm 1 , the mean free path in the near IR is 1 mm.
Diffusion theory approximation applies at source-to-detector distances much
larger than 1mm, typically 1cm and up to 7 cm. Measurement at a great
distance from the source is referred to as the far field condition. Mean free paths
between interaction sites between photons and tissue typically range from 0.01
mm to 2 mm, with 0.75 mm being a typical value in the visible spectrum. In the

WO 99/59464 PCT/US99/10812

present invention, the detector distances can be as short as 0.4 mm to 5 mm.
Thus, they are either shorter than or of a comparable value to the mean free
path, and the diffusion theory approximation does not hold. Measurements at
small separations of source and detector present a near field condition.

5 A more exact solution of the light transport equation in turbid media can

be obtained by following the path of each individual photon and calculating the
probability of scattering and/or absorption in a series of steps using Monte Carlo
simulation. Physical quantities of interest are scored within statistical
uncertainties of the finite number of photons simulated. The power of the Monte

10 Carlo method lies in its ability to handle virtually any source, detector, and tissue
boundary condition, as well as any combination of optical properties of tissue.
Monte Carlo methods can also accommodate polarized light and diffraction
effects in the light propagation calculation, and these methods are preferred in
the present invention over the diffusion theory approximation.

15 Monte Carlo simulations were used for the probe geometry of the present

invention. The public domain software program employed was " Monte Carlo
simulations of mufti-layered turbid media", by Lihong Wang and Steven L
Jacques, obtained from Oregon Laser Center, Portland, Oregon.

In the Monte Carlo model, the beam diameter was 400 micrometers, the

20 number of photons injected was 200,000 per run, light was propagating from
fiber (n = 1 .5) into tissue (n = 1 .4). The thickness of the tissue layer was set from
5 to 25 mm. Light reflected at the 0.44 mm, 0.78 mm, 0.89 mm, 1.17 mm, 1.35
mm, and 1.81 mm distances from the point the light was injected were calculated
for a matrix of several m* and n a values. These distances corresponded

25 approximately to the positions of fibers 23, 24, 25, 26, 27, and 28. The resultant
•°9e RO) vs log e (R/Rj) were plotted as a grid. The constant p a and p 5 points were
connected to form a grid in the log e R(i) vs log e (R/Rj) space, where R(i) ^
represents reflectance at a distance i and R(j) represents reflectance at a
distance j. Spatially resolved backscattering was determined for a set of

30 IntralipkJ solutions, hemoglobin solution in Intralipid suspension, opal glass, and

WO 99/59464

PCT/US99/10812

plastic rods polymerized to incorporate different levels of scatter and absorbing
pigment. The experimental values were overlapped on the Monte Carlo-
generated grid, and absorption and scattering coefficients of the reference
material were determined by the use of tables generated from the grid. Spatially
5 resolved light scattering (SRLC) measurements were obtained on the dorsal part
of the forearms of human volunteers and plotted on the same graph. An
example of the result of Monte Carlo simulation and the measured reflectance for
a human volunteer are shown in FIG. 9. The dashed lines connecting the circles
represent experimental data. The solid lines represent Monte Carlo fits for the

10 absorption and scattering coefficients indicated. The graph indicates that the
reflectance values predicted by the model are close to the experimental results.
H 8 ' and n a values for several Caucasian, Oriental and Mediterranean subjects
were determined at 34° C. The average values of m' and ^ at several
illumination wavelengths were used to calculate the mean free path (mfp') and

15 are shown in Table 1.

WO 99/59464 PCT/US99/10812

Table 1

Average optical constants and mean free path for human subjects

Optical
constant

550 nm

590 nm

650 nm

750 nm

800 nm

900 nm

W + Ha)

(cm 1 )

Mean free
path (mm)

0.62

0.72

0.88

1.03

1.1

1.23

Penetration
depth (mm)

0.72

0.92

1.42

1.67

1.92

2.04

5 Thus the measured mean free path is of the same magnitude as the separation
of the source from the detector, thereby justifying the use of Monte Carlo
modeling. The penetration depths achieved were less than or equal to 2 mm.
The majority of the reflected light sampled at depths in the skin less than or
equal to about 2 mm. Other longer wavelengths up to 2500 nm can be selected

10 to achieve shallow penetration depth.

The effect of changes of temperature on the scattering and absorption
coefficients of a diabetic and a non-diabetic individual were tested by means of
the SRLS apparatus described in Example 1, and absorption and scattering
coefficients were determined from the Monte Carlo-generated grid. Temperature

15 of the tissue was varied from 20° C to 45° C. Concentrations of glucose and
hemoglobin in blopd were measured by means of a commercial instrument
(Vision®, Abbott Laboratories) prior to the SRLS measurement Glucose
concentration in the nondiabetic subject was 88 mg/dL and glucose
concentration in the diabetic subject was 274.6 mg/dL SRLS measurements

20 were performed on the forearm of each subject.

The reduced scattering coefficient increased with increased temperature
at all wavelengths for the two subjects. d^VdT ranged from 0.044 to 0.0946 for

WO 99/59464

PCT/US99/10812

the non-diabetic subject and from 0.0633 to 0.0881 cm 1 / °C for the diabetic
subject. The change in the refractive index of water over the same temperature
range was approximately -IxlO" 4 per °C. The change in the reduced scattering
coefficient for 1000 nm spherical particles over the same temperature range is
calculated using the equation of Graaf et al (Equation (3)) to be 0.024 cm- 1 per
°C. Thus the measured dm'/dT for the forearm of the test subjects is larger than
the calculated values for 1000 nm particles, which mimic the biological tissue.
Dependence of the scattering coefficient (in tissue) on temperature is greater
than dependence of the scattering coefficient (in 1 mm particles) on temperature,
which, in turn, is much greater than dependence of refractive index on
temperature.

FIG. 1 0 shows the spectral distribution of the derivative of scattering
coefficient with respect to temperature (dm'/dT) for the two subjects. A
difference in the spectral response between the two subjects can be observed.
The fractional change in the derivative is shown in FIG. 11. A noticeable
difference between the two subjects was observed, especially at the non-
absorbing wavelengths (away from the visible hemoglobin absorption bands) •

The derivative of the absorption coefficient with respect to temperature is
shown in FIG. 12 for the two subjects. The spectral distribution of the derivative
du^dT differs between the two temperature ranges of 20 to 35° C, and 35 to 40°
C. At temperatures below 35° C (the ambient skin temperature), slight change in
the dpa/dT as a function of temperature can be observed. However, the values
differed for the two subjects. At temperatures above 35° C, the absorption
derivative at the hemoglobin visible absorbing wavelengths is much higher than
that at the non-absorbing wavelengths, which suggests a change in blood
perfusion. The shape of the curve is similar to that of hemoglobin absorption.
Blood perfusion to the skin at higher temperature may account for this similarity.
There is an observed difference between diabetic and non-diabetic subjects.

WO 99/59464 PCT/US99/10812

Various modifications and alterations of this invention will become
apparent to those skilled in the art without departing from the scope and spirit of
this invention. It should be understood that this invention is not to be unduly
limited to the illustrative embodiments set forth herein.

5 For example, while in-vivo glucose measurement has been illustrated,

other measurements, in-vivo or in-vitro, needing improved specificity could
benefit from measurements with combined technologies (i., e., alcohol, blood
urea nitrogen (BUN), bilirubin, hemoglobin, creatine, electrolytes, blood gases,
and cholesterol). It should be recognized that the wavelengths used for

10 measurement may vary for the different analytes of interest.

A variety of detectors may be employed in the present invention without
departing from the spirit of the invention. Preferably, the detectors should be
optimized for the particular measurement to be made, with wavelength, cost,
performance, and engineering design being considered. The detectors may be

15 either single element or array detectors. While single element detectors are
generally less costly and more amenable to frequency modulation and detection
schemes, an alternative embodiment could use detector arrays, such as a
photodiode array or a charge-coupled device (CCD) array, for multi-wavelength
detection.

20 Various filters and the like that transmit only the wavelength(s) of interest

may be placed in front of the detectors. Such filters may include, for example,
dielectric filters, holographic filters, and tunable filters, such as an Acoustp-Optic
Tunable Filter (AOTF). Alternatively, frequency modulation may be used to
distinguish the one measured signal from another. The development of

25 detectors having sensitivities extending continuously from visible wavelengths
into the infrared region will permit the use of a single detector, or detector array,
over a large spectral range, without the need to switch detectors.

Although the optical detection method used in the examples is spatially-
resolved diffuse reflectance, other methods that can lead to calculating the

30 absorption and scattering coefficients of a turbid medium can be used by those

WO 99/59464

PCT/US99/10812

skilled in the art. thus any optical measurement that allows control of
temperature over an area larger than the area of optical measurement can be
used. An example of such a measurement is diffuse reflectance using
randomized optical fiber bundles. Another example involvesfrequency
5 modulation measurements using a high enough modulation frequency to allow
measuring a phase angle change over a small separation of source and
detector. Yet another modification would be the use of polarimetric
measurements utilizing polarization-conserving fibers. Other methods of
calculation can be used, such as neural networks and data mining

10 methodologies.

For non-invasive measurements on a body part, the body interface
module may be adapted to change the shape of the body part or to change the
physical relationship between the transducers and the body part. For example,
the body interface module might be adapted to increase the pressure applied to

15 the body part by the transducer. Such a change might be made, for example, to
alter local perfusion rates.

WO 99/59464

PCT/US99/I0812

What is claimed is:

1 . A method of determining at least one parameter of a body part at a
5 given temperature comprising the steps of:

(a) modulating said temperature of said body part;

(b) measuring at least one optical property of said body part, at at least
one wavelength, as a function of said temperature modulation, to obtain an
optical measurement, wherein the volume of the body part subject to

10 temperature modulation is greater than the volume of the body part that is being
measured in the optical measurement; and

15 2. The method of claim 1 , wherein said at least one parameter is the

presence of an analyte.

3. The method of claim 1 , wherein said at least one parameter is the
concentration of an analyte.

4. The method of claim 1 wherein said at least one parameter is the
presence of a tissue heterogeneity.

5. The method of claim 1 , wherein said at least one parameter is a
25 change in blood circulation.

6. The method of claim 1, wherein said at least one
optical property is a scattering coefficient.

WO 99/59464

PCT/US99/I0812

7. The method of claim 1 , wherein said at least one optical property is
an absorption coefficient.

8. The method of claim 1 , wherein said optical measurement is
5 performed at two or more wavelengths.

9. The method of claim 1 , wherein the volume of the body part subject
to temperature modulation is at least two times the volume of the body part being
optically sampled.

10. The method of claim 1 , wherein the volume of the body part subject
to temperature modulation is at least five times the volume of the body part being
optically sampled.

15 11 - The method of claim 1, wherein said method of correlation is

selected from the group consisting of least squares, partial least squares, and
neural networks.

1 2. The method of claim 1 , wherein said wavelength ranges from 400
20 nm to 2500 nm.

13. A method of measuring at least one parameter of a body part at a
given temperature, comprising the steps of:

(a) decreasing the temperature of said body part to a temperature
25 which is at or below the normal physiological temperature of said body part;

(b) determining at least one optical property of said body part at said
temperature of step (a);

(c) increasing the temperature of said body part to a temperature
above the normal physiological temperature of said body part;

WO 99/59464

PCT/US99/10812

(d) determining at least one optical property of said body part at said
temperature of step (c); and

(e) analyzing the optical properties determined in steps (b) and (d) to
obtain a measurement of said at least one parameter of said body part.

14. The method of claim 1 3, further comprising the step of measuring
scattering coefficient as a function of temperature.

1 5. The method of claim 1 3, further comprising the step of measuring
10 absorption coefficient as a function of temperature.

16. The method of claim 13, further comprising the step of correlating
the measured optical property with concentration of an analyte in the body, said
concentration determined by a method independent of steps (a), (b), (c), and (d).

15 : •

1 7. The method of claim 1 3, wherein said parameter is the presence of
an analyte.

18. The method of claim 13, wherein said parameter is the
20 concentration of an analyte.

1 9. The method of claim 1 3 wherein said parameter is the presence of
a tissue heterogeneity.

25 20. The method of claim 1 3, wherein said parameter is the presence of

a blood circulation change.

21 . The method of claim 1 3, wherein said method of correlation is
selected from the group consisting of least squares, partial least squares, and
30 neural networks.

WO 99/59464

PCT/US99/10812

22. The method of claim 1, wherein said wavelength ranges from 400
nm to 2500 nm.

23. A method of measuring at least one parameter of a body part,
comprising the steps of:

(a) adjusting the temperature of said body part to a temperature that is
substantially the same as the norma! physiological temperature of said body part;

(b) determining at least one optical property of said body part at said
temperature of step (a);

(c) reducing the temperature of said body part to a temperature that is
lower than the normal physiological temperature of said body part;

(d) determining at least one optical property of said body part at said
temperature of step (c);

(e) increasing the temperature of said body part to above the normal
physiological temperature of said body part;

(f) . determining at least one optical property of said body part at said
temperature of step (e); and

(g) analyzing the measurements of steps (b), (d), and (f) to obtain a
measurement of said at least one parameter of said body part.

24. Method of claim 23, further comprising the step of correlating the
measured optical property with concentration of an analyte in the body, said
concentration determined by a method independent of steps (a), (b), (c), (d), (e),
(f). and (g).

25. The method of claim 23, wherein said parameter is the presence of
an analyte.

WO 99/59464

PCT/US99/10812

26. The method of claim 23, wherein said parameter is the
concentration of an analyte.

27. The method of claim 23, wherein said parameter is the presence of
5 a tissue heterogeneity.

28. The method of claim 23, wherein said parameter is the presence of
a vascular change.

10 29. The method of claim 23, wherein said correlation method is

selected from the group consisting of least squares, partial least squares, and
neural networks.

30. The method of claim 23, wherein the volume of the body part

1 5 subject to temperature modulation is at least two times the volume of the body
part being optically sampled.

31 . The method of claim 23, wherein the volume of the body part
subject to temperature modulation is at least five times the volume of the body

20 part being optically sampled.

32. The method of claim 1 , wherein said wavelength ranges from 400
nmto2500nm.

25 33. An apparatus for measuring concentration of an analyte in a body

part comprising:

(a) a temperature controlling element adapted to conform to the surface of
said body part;

(b) at least one light transmitting element and at least one light receiving
30 element located within said temperature controlling element (a);

WO 99/59464

PCT/US99/10812

(c) at least one light source and at least one detector to illuminate a
defined volume of the body part subject to temperature control; and

(d) a signal processor to determine an optical property of the body part,
said temperature controlling element capable of controlling the temperature of a

5 volumetric portion of the body part that is larger than the volumetric portion of the
body part being illuminated by the at least one light source and at least one
detector.

10 34. The apparatus of claim 33 where the distance of the source from

the detector and the wavelengths of the source are selected to limit the depth of
penetration in the tissue to a that wherein the temperature is being controlled,

35. The apparatus of claim 34, wherein said wavelength ranges from
15 400 nm to 2500 nm.

36. The apparatus of claim 33 where the wavelength of the light
sources ranges from 600 nm to 1300 nm.

20 37. The apparatus of claim 33 where the separation distance between

center of light emitting and receiving elements is no more than 6 mm.

38. The apparatus of claim 33, wherein the depth in said body part is
confined to a depth no greater than 2 mm.

WO 99/59464

1 / 9

PCT/US99/10812

DETECTOR 1

SOURCE

DETECTOR 3

DETECTOR 2

dav3

LIGHT
SOURCE
MODULE

DETECTOR
MODULE

-18

HUMAN
INTERFACE
MODULE

FIG. 2

WO 99/59464

2/9

PCT/US99/I0812

WO 99/59464

3/9

PCT/US99/10812

WO 99/59464

4/9

PCT/US99/10812

WO 99/59464

5/9

PCT/US99/10812

30.4°C DISK

FIG. 7C

WO 99/59464

6/9

PCT/US99/I0812

WO 99/59464

PCT/US99/10812

7/9

WO 99/59464

8/9

PCT/US99/10812

WO 99/59464

9/9

PCT/US99/10812

INTERNATIONAL SEARCH REPORT

Inter, jnai Appiicatton No

PCT/US 99/10812

A. CLASSIFICATION OF SUBJECT MATTER

IPC 6 A61B5/00 G01N21/47

According to International Patent Classification (IPC) or to both national classification and IPC

B. FIELDS SEARCHED

Minimum documentation searched (classification system followed by classification symbols)

IPC 6 A61B G01N

Documentation searched other than minimum documentation to the extent that such documents are Incl

udod In the fields searched

Electronic data base consulted during the international search (name of data base and, where practical, search terms used)

C. DOCUMENTS CONSIDERED TO BE RELEVANT

Category *

Citation of document with indication, where appropriate, of the relevant passages

Relevant to claim No.

WO 98 03847 A (HILLS ALEXANDER K)
29 January 1998 (1998-01-29)

page 1, line 4 - line 14

page 3, line 25 - line 29

page 7, line 29 - page 8, line 19

page 10, line 20 - page 11, line 5

page 11, line 23 - line 30

page 12, line 19 - line 22

page 12, line 23 - page 13, line 32

page 19, line 14 - line 22

page 20, line 4 - Hne 9

1-3,

7-10,

12-18,

22-26,

30-33,

35,36

4-6,11,

19-21,

27-29,

34,37,38

Further documents are listed in the continuation of box C.

Patent family members are listed in annex

* Special categories of cited documents :

"A" document defining the general state of the art which is not

considered to be of particular relevance
"E" earlier document but published on or after the international

filing date

X* document which may throw doubts on priority clalm(s) or
which Is cited to establish the publication date of archer
citation or other special reason (as specified)

"O" document referring to an oral disclosure, use, exhfcitionor
other means

"P" document published prior to the international filing date but
later than the priority date claimed

T later document published after the international filing date
or priority date and not in conflict with the application but
cited to understand the principle or theory underlying the
Invention

"X* document of particular relevance; the claimed Invention
cannot be considered novel or cannot be considered to
involve an Inventive step when the document is taken alone

"Y* document of particular relevance; the claimed Invention
cannot be considered to involve an inventive step when the
document is combined with one or more other such docu-
ments, such combination being obvious to a person skiSed
in the art

*&* document member of the same patent family

Date of the actual completion of the international search

16 July 1999

Date of mailing of the international search report

26/07/1999

Name and mailing address of the ISA

European Patent Office, PB. 581 8 Patentfaan 2
ML - 2260 HV Rfjswn*
Tel. (431-70) 340^040, Tx. 31 651 epo nt.
Fax (+31-70) 340-3016

Fomi PC771S/V210 (second shoot) (Juty 1992) ~~

Authorized officer

Navas Montero, E

page 1 of 2

INTERNATIONAL SEARCH REPORT

In ton. _»nal Application No

PCT/US 99/10812

(^Continuation) DOCUMENTS CONSIDERED TO BE RELEVANT

Category • Citation ot document, with indicatk>awnefe appropriate, of the relevant passages

Relevant to daim No.

page 21, line 20 - line 25
page 20, line 24 - line 28
claims 1,2,11,12
figures 5,6

US 5 672 875 A (BLOCK MYRON J ET AL)
30 September 1997 (1997-09-30)

column 1, line 28 - line 38

column 7, line 6 - line 15

figures 1,3

DE 196 34 152 A (SIEMENS AG)
5 March 1998 (1998-03-05)

column 1, line 41 - line 46

column 1, line 48 - line 58

column 1, Hne 65 - column 2, line 11

claims 1,3-5,7; figures 1,2

WO 95 20757 A (MINNESOTA MINING & MFG)
3 August 1995 (1995-08-03)

page 17, line 3 -.line 8

figure 2; table 1

DE 44 17 639 A (B0EHRINGER MANNHEIM GMBH)
23 November 1995 (1995-11-23)

column 4, line .42 - line 49

column 6, Hne 15 - line 30

US 5 131 391 A (SAKAI HIROSHI- ET AL)

21 July 1992 (1992-07-21)
column 1, Hne 31 - line 43
column 2, Hne 47 - line 57; figures

1,2,5

4,6,19,
27

5,20,28,
37,38

11,21,29

FocnPCT/TSA^10(conttnus4oncJ»«oodsh^(JcV

page 2 of 2

INTERNATIONAL SEARCH REPORT

Information on patent family members

Inter, jnal Application No

PCT/US 99/10812

Patent document

Publication

Patent family

Publication

caea in search report

udltf

member(8)

date

WO 9803847

29-01-1998

All

3721997 A

10-02-1998

US 5672875

30-09-1997

5818048 A

06-10-1998

5424545 A

13-06-1995

5434412 A

18-07-1995

5382696 A

30-12-1996

2223408 A

19-12-1996

0884970 A

23-12-1998

Q639922 A

19-12-1996

9614567 A

17-05-1996

20-07-1995

0742897 A

u# ttu? f n

20-11-1996

.IP

Q510884 T

7 jIwOO't 1

04-11-1997

^awiJUC. n

20-07-1998

All
nu

26-03-1998

7842894 A

01-05-1995

2173200 A

13-04-1995

0721579 A

9503585 T

08-04-1997

9510038 A

io— v*i-iyyD

IIC
Uo

CQ1Qn/A A
OOIOvHH n

06-10-1998

30/0070 n

26-11-1996

5543459 A

06-08-1996

DE 19634152

05-03-1998

QA0ftn7fi A

jOUOU/ U n

« M 1 OftQ

2O-0Z-1998

WO 9520757

03-08-1995

lie

10-09-1996

07d?ftQfi A
U/HcO^O n

20-11-1996

.IP
ur

Q50RPQ1 T

26-08-1997

5755226 A

26-05-1998

DE 4417639

oo_i i _i one

2342595 A

18-1 ?-1QQ>>

9532416 A

30-11-1995

19580537 0

01-04-1999

0760091 A

OR— nt-10Q7

UO^IO 177/

10500338 T

13-01-1998

5770454 A

23-06-1998

US 5131391

21-07-1992

2766317 B

18-06-1998

3023846 A

31-01-1991

floim PCT/ISACtO (patent twnty annaX) (July 1992)

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "USPTO Patents Application 10687850"

See other formats